One of the problems involved in microwave spectroscopy is that of containing a sample material, having pressure-dependent spectral lines, within a microwave cavity field at hydrostatic pressures of several thousand atmospheres.
Previous methods of attaining extremely high pressures inside a microwave cavity have involved making the cavity wall itself the pressure vessel or container. When this is done, the walls of the cavity have to be made of nonmagnetic, conducting materials of high tensile strength. For example, the cavity was machined from a block of beryllium-copper alloy and the hydrostatic medium filled the entire cavity. The microwaves were coupled into the cavity through the cavity walls by an antenna connected to an electrical feed-through in an insulating plug or bushing. This method had the following disadvantages:
A. The necessity of machining the microwave cavity; PA1 B. The sensitivity of the microwave detection system was reduced because of: PA1 C. Safety hazards accompanied by attainment of such high pressure in such a large volume (e.g. 30 cm.sup.3) of highly compressed fluid.
1. Microwave losses at the feed-through; PA2 2. Increased pole separation of the electromagnetic poles necessary to accomodate the greatly enlarged cavity; and PA2 3. Detuning of the cavity frequency due to pressure-induced cavity deformations; and
An object of the present invention is to provide a separate high pressure sample container that can be inserted or positioned within the microwave field of a standard cavity resonator or waveguide. In accordance with the invention, the sample container comprises an elongated dielectric tube capable of withstanding an internal pressure of at least 2000 atmospheres, and having a dielectric constant not greater than 4, whereby it is transparent to microwaves. The outer diameter of the dielectric tube is at least five times its inner diameter, to withstand such a high pressure. The dielectric tube has an open end connected by a high pressure seal to a source of the desired pressure, and a closed end which is positioned within a microwave cavity or waveguide. A preferred dielectric material, which has been successfully used for this purpose, is nylon, with a dielectric constant of 3.5 at 10.sup.6 cycles/sec. Another suitable material is Delrin, with a dielectric constant of 3.7 at 10.sup.6 cycles/sec.
The ultimate tensile strength of nylon is in the range of 12-14.times.10.sup.3 psi, which would appear to limit the use of nylon as the pressure tube material to 1-1.5.times.10.sup.3 atmospheres. However, the ultimate tensile strength of a material is not necessarily the upper limit of its ability to contain internal pressure. Consider a hollow cylinder of inside radius a and outside radius b. The tangential stress at any radius r where a.ltoreq.r.ltoreq.b is given by ##EQU1## where P.sub.i is the internal pressure and w=b/a. At the inner surface (r=a) this reduces to ##EQU2## which implies that .sup.S t&gt;P.sub.i. However, at the outer surface (r=b), the formula reduces to ##EQU3## which, for a 5:1 wall ratio as chosen, implies that .sup.S t&lt;1/10 P.sub.i. Accordingly, a material having a tensile strength of 10.sup.4 psi could contain pressures approaching 10.sup.5 psi, or 7000 atmospheres, even though the inner surface is stressed beyond its elastic limit. The normal strengths are true only for materials in the elastic state, a condition that does not exist in the high pressure dielectric tube used in the present invention. Here, however, is the beauty of a material like nylon which has an enormous elongation (e.g. 15%) at the yield point. This means that although the stresses exceed the tensile strength at the inner surface, the material remains elastic because it cannot stretch beyond its yield point--there is no place for the material to go. It is confined all around on the outside by material that is stressed to values less than the ultimate strength thereof. As usual, the theoretical limit of this idealized condition is not attainable experimentally. Pressures of 7000 atmospheres have not yet been achieved. However, tests have shown that the overstressed material near the inner surface of the dielectric sample tube does exhibit unusual mechanical strength. There is a pressure gradient in which the stress decreases as r goes from a to b. The area of a virtual concentric surface of radius r increases as r.sup.2, and since the applied radial force is at most constant (if it is not reduced by the strength of the interior overstressed material) the pressure falls off at least as r.sup.2, so that at some point P.sub.i gets small enough that the outer material can contain it (assuming that there is still some outer material left, i.e., r&lt;b).